Automatic traffic event detection method

Abstract

The invention relates to an automatic traffic event detection method. The automatic traffic event detection method is characterized by comprising steps that a to-be-detected regional road network is divided into multiple road segments, and traffic detectors are distributed upstream and downstream portions of each road segment P; original traffic parameter data is acquired by the traffic detectors according to a sampling period and is uploaded through a communication system to a traffic information center data storage device; rolling time sequence synthesis of the original traffic parameter data acquired by the upstream and downstream traffic detectors at the P road segment at an ith period is carried out by a traffic information center platform according to a determined time scale. The method is advantaged in that through rolling time sequence synthesis of the original traffic parameter data, a problem that rolling time sequence synthesis of the original traffic parameter data is carried out at a fixed time scale in a traffic event detection method in the prior art is solved, in combination with the upstream and downstream detector speed and the occupancy information, a traffic flow operation state can be more actually reflected.

Description

technical field [0001] The invention relates to a traffic event detection method in the technical field of intelligent transportation, in particular to a traffic event automatic detection method based on rolling time series. Background technique [0002] Traffic incidents include traffic accidents, scattered goods, vehicle breakdowns, etc., which will not only lead to a large number of traffic delays, but also cause environmental pollution and decrease in road traffic safety. According to the relevant statistics of the U.S. Federal Highway Administration, the traffic congestion caused by traffic incidents is as high as 70% of the total congestion. Every minute a traffic incident lasts, it will cause an average delay of at least 4 to 5 minutes for each vehicle. The ratio will be much higher. Therefore, accurate and timely traffic event detection is of great significance for alleviating traffic congestion. [0003] The essence of Automatic Incident Detection (AID) method is ...

AI Technical Summary

Problems solved by technology

Although the synthesis of a fixed time scale will reduce the false alarm rate of the AID method to a certain extent, with the increase of the synthesis time scale, the average detection time will be extended and the detection rate will be reduced, that is, the simultaneous improvement of the event detection performance evaluation index cannot be guaranteed.

Images